Understanding the process:
I read an article about geology (In Modeling, Simplicity isn’t Simple, Chris Paola NATURE vol. 469 no. 7328 January 6, 2011 page 38) describing efforts to create computer models of geological phenomena like the way stream beds and watersheds evolve. 

Computer modeling has been very much on my mind recently.  Some years ago I prepared a computer program that used a post-zygotic model to predict with some success the way population size and fertility interact.  The program is not perfect.  There is data I have at hand that does not quite fit the model.  So I have undertaken the pre-zygotic approach.

A zygote is Greek for a yoke of oxen.  In biology a zygote is a fertilized ovum.  Infertility caused by something that happens before the egg and sperm can unite is called pre zygotic.  Infertility caused by subsequent events is post zygotic.  In my mind it is a sort of question like, “Which came first, the chicken or the egg?”  Since the process cycles whether something is before or after the zygote depends on which zygote.  But the computer program has no such qualms.  To write the program you have to go one way or the other or, as I plan, to have them both going on at the same time.

The crashes have been spectacular.  So far every disaster has been traced to a flaw in the programming language I have written.  I have used C++, which is no longer supported by Microsoft but which seems the best for my needs.  It has its quirks.  Do programmers use obscene language?  They do if they use C++.  It is inherently obscene. 

The process has given me the cold creepy crawlies in my skin.  I can’t eat, and I’m always hungry.  That’s good.  When I started I also got the dry heaves, nosebleeds, couldn’t sleep and couldn’t stay awake.  As I remember there were more things than that, but I have managed to drive them beyond recall.

Maybe the thing will work this time around.

So the article contains a remark by Paola that a complex computer model entails a risk of having a program you do not understand that models a process you do not understand.  True words they are.  After all, if I really understood the program I could get it right.  I suppose it is no comfort to know others have, on a much loftier scale of course, the problems one is having oneself.  But if I can get it to function in a reasonable way then I shall smugly conclude that I knew what I was doing all along.

There have been 10,728 visitors so far.

Home page.